aiox-core 5.0.5 → 5.0.7

package/pro/squads/squad-creator/docs/sop-extraction-process.md

@@ -0,0 +1,675 @@
+# SOP Extraction Process Documentation
+
+> **Pattern:** SC-DOC-001 (Process Documentation)
+> **Version:** 1.0.0
+> **Squad:** squad-creator
+> **Last Updated:** 2025-01-24
+
+---
+
+## Overview
+
+The SOP Extraction Process transforms meeting transcripts into structured, automation-ready Standard Operating Procedures (SOPs) that can be immediately converted into working AIOX Squads.
+
+### Purpose
+
+1. **Capture tacit knowledge** - Extract implicit process knowledge from recorded meetings
+2. **Structure for automation** - Classify each step by cognitive type and automation potential
+3. **Enable hybrid execution** - Define where humans and AI collaborate
+4. **Generate working squads** - Produce blueprints ready for immediate squad creation
+
+### Key Patterns Used
+
+| Pattern | Name | Purpose |
+|---------|------|---------|
+| SC-PE-001 | SOP Extraction Standard | 11-part SOP structure |
+| SC-CK-001 | SOP Validation Checklist | Quality gates before handoff |
+| HO-TP-001 | Task Anatomy | 8-field task structure |
+| HO-EP-001-004 | Executor Types | Human/Agent/Hybrid/Worker |
+| PV_PM_001 | Automation Tipping Point | When to automate decision |
+
+---
+
+## Pipeline Overview
+
+```
+┌─────────────┐     ┌─────────────┐     ┌─────────────┐     ┌─────────────┐
+│ Transcript  │ ──► │ @sop-       │ ──► │ Validation  │ ──► │ @squad- │
+│ (any source)│     │  extractor  │     │ (SC-CK-001) │     │  creator    │
+└─────────────┘     └─────────────┘     └─────────────┘     └─────────────┘
+                           │                   │                   │
+                           ▼                   ▼                   ▼
+                    ┌─────────────┐     ┌─────────────┐     ┌─────────────┐
+                    │ SOP Doc     │     │ Human       │     │ Working     │
+                    │ (11 parts)  │     │ Review      │     │ Squad       │
+                    └─────────────┘     └─────────────┘     └─────────────┘
+```
+
+---
+
+## Detailed Process Flow
+
+### Phase 1: Data Source Resolution
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 0: DATA SOURCE RESOLUTION                                   │
+│  ─────────────────────────────────────────────────────────────── │
+│  • Lê config/squad-config.yaml                                    │
+│  • active_source = supabase | local_file | api | direct           │
+│  • Busca transcrição e normaliza para schema padrão               │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**Configuration:** `config/squad-config.yaml`
+
+```yaml
+data_sources:
+  transcripts:
+    active_source: supabase  # ← change this to switch sources
+
+    sources:
+      supabase:
+        type: database
+        query:
+          table: transcripts
+          filters:
+            - field: status
+              operator: eq
+              value: ready
+
+      local_file:
+        type: file
+        config:
+          base_path: "inputs/transcripts/"
+
+      api:
+        type: http
+        connection:
+          base_url_env: TRANSCRIPT_API_URL
+
+      direct:
+        type: inline
+        # Transcript passed as parameter
+```
+
+**Output Schema:**
+
+```yaml
+transcript:
+  transcript_id: string       # Required
+  transcript_content: string  # Required
+  transcript_source: string   # Optional: "meeting", "interview", "audio"
+  transcript_url: string      # Optional: original source
+  transcript_duration: number # Optional: seconds
+  transcript_participants: [] # Optional: speaker names
+  transcript_date: date       # Optional: when recorded
+  transcript_language: string # Optional: ISO code
+  transcript_metadata: {}     # Optional: custom data
+```
+
+---
+
+### Phase 2: Structure Extraction (First Pass)
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 1-2: FIRST PASS - STRUCTURE EXTRACTION                      │
+│  ─────────────────────────────────────────────────────────────── │
+│  • Identifica: nome do processo, objetivo                         │
+│  • Lista todos os passos mencionados                              │
+│  • Captura: pessoas, ferramentas, sistemas                        │
+│  • Marca pontos de decisão ("se", "quando", "depende")            │
+│                                                                   │
+│  OUTPUT: Raw extraction notes                                     │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**Transcript Signals to Detect:**
+
+| Signal Type | Keywords | Example |
+|-------------|----------|---------|
+| **Sequence** | "first", "then", "after", "next", "finally" | "First I open the system, then I check..." |
+| **Decision** | "if", "when", "depends", "unless", "otherwise" | "If the client is premium, we prioritize..." |
+| **Precondition** | "before", "must have", "requires", "need to" | "Before starting, I need access to..." |
+| **Heuristic** | "usually", "generally", "most of the time" | "Usually we wait 24 hours..." |
+| **Exception** | "except", "unless", "but if", "special case" | "Except when it's urgent..." |
+| **Implicit** | "always do", "we just", "obviously" | "We obviously check the CRM first..." |
+
+**Red Flags to Mark:**
+
+| Red Flag | Meaning | Action |
+|----------|---------|--------|
+| "depends on who does it" | Non-standardized variation | Document variation, flag for standardization |
+| "we figure it out" | Undocumented exception | Add to gaps, request clarification |
+| "[name] knows how" | Single point of failure | Flag as risk, document knowledge |
+| "this is rare" | Exception becoming rule | Verify frequency, may need process |
+| "always been this way" | Potentially obsolete | Question necessity, validate |
+
+---
+
+### Phase 3: Task Anatomy Mapping
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 3: TASK ANATOMY MAPPING (HO-TP-001)                         │
+│  ─────────────────────────────────────────────────────────────── │
+│  Para CADA passo extraído, preencher 8 campos obrigatórios:       │
+│                                                                   │
+│  ┌─────────────────────────────────────────────────────────────┐ │
+│  │ task_name: "Verificar status do cliente"                    │ │
+│  │ responsible_executor: @customer-service                     │ │
+│  │ execution_type: Hybrid                                      │ │
+│  │ estimated_time: 5m                                          │ │
+│  │ input: [customer_id, crm_access]                            │ │
+│  │ output: [customer_status, risk_flag]                        │ │
+│  │ action_items: [1. Abrir CRM, 2. Buscar cliente, 3. Avaliar] │ │
+│  │ acceptance_criteria: [Status documentado, Flag definido]    │ │
+│  └─────────────────────────────────────────────────────────────┘ │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**Task Anatomy Fields (HO-TP-001):**
+
+| Field | Type | Required | Description |
+|-------|------|----------|-------------|
+| `task_name` | string | ✓ | Verb + Object format (e.g., "Verify customer status") |
+| `status` | enum | ✓ | pending, in_progress, review, completed, blocked |
+| `responsible_executor` | string | ✓ | @agent reference or role name |
+| `execution_type` | enum | ✓ | Human, Agent, Hybrid, Worker |
+| `estimated_time` | duration | ✓ | With unit: 5m, 2h, 1d |
+| `input` | array | ✓ | ≥1 item required |
+| `output` | array | ✓ | ≥1 item required |
+| `action_items` | array | ✓ | Atomic steps to complete task |
+| `acceptance_criteria` | array | ✓ | How to verify success |
+
+---
+
+### Phase 4: Cognitive Classification
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 3b: COGNITIVE CLASSIFICATION                                │
+│  ─────────────────────────────────────────────────────────────── │
+│                                                                   │
+│   "Eu olho no sistema..."        → Perception    → ✅ Agent       │
+│   "Eu decido se vale a pena..."  → Judgment      → ⚠️ Hybrid      │
+│   "Eu ligo pro cliente..."       → Negotiation   → ❌ Human       │
+│   "Eu copio os dados..."         → Memory        → ✅ Worker      │
+│                                                                   │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**Cognitive Taxonomy:**
+
+| Type | Description | Automatable? | Executor |
+|------|-------------|--------------|----------|
+| **Perception** | Identify patterns in data | ✅ Yes | Agent (vision, NLP) |
+| **Memory/Retrieval** | Fetch known information | ✅ Yes | Worker (query) |
+| **Analysis** | Decompose, compare, evaluate | ✅ Yes | Agent |
+| **Synthesis** | Combine information into new whole | ✅ Yes | Agent |
+| **Judgment** | Decide with incomplete information | ⚠️ Partial | Hybrid |
+| **Creativity** | Generate genuinely novel output | ⚠️ Partial | Hybrid |
+| **Empathy** | Understand others' emotional state | ❌ No | Human |
+| **Negotiation** | Influence others' decisions | ❌ No | Human |
+| **Accountability** | Assume consequences | ❌ No | Human |
+| **Ethics** | Decide between conflicting values | ❌ No | Human |
+
+**Executor Assignment Rules:**
+
+| Executor | When to Use | Examples |
+|----------|-------------|----------|
+| **Human** | Irreversible decisions, negotiation, empathy, accountability | Fire someone, negotiate contract, handle complaint |
+| **Agent** | Large volume analysis, content generation, classification, patterns | Summarize 100 docs, categorize leads, draft responses |
+| **Hybrid** | Medium risk, needs human "sanity check" | Draft important email (AI writes, human reviews) |
+| **Worker** | 100% deterministic, no ambiguity, repetitive | Move file, call API, format date, send notification |
+
+---
+
+### Phase 5: Decision Rules Extraction
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 4: DECISION RULES EXTRACTION                                │
+│  ─────────────────────────────────────────────────────────────── │
+│                                                                   │
+│  Transcrição: "Se o cliente for premium, a gente prioriza..."     │
+│                           │                                       │
+│                           ▼                                       │
+│  ┌──────────────────────────────────────────────────────────┐    │
+│  │ IF customer.tier == "premium"                             │    │
+│  │ THEN priority = HIGH                                      │    │
+│  │ ELSE priority = NORMAL                                    │    │
+│  │                                                           │    │
+│  │ Automatable? ✅ Yes (deterministic rule)                  │    │
+│  └──────────────────────────────────────────────────────────┘    │
+│                                                                   │
+│  Transcrição: "Depende do feeling, às vezes a gente..."           │
+│                           │                                       │
+│                           ▼                                       │
+│  ┌──────────────────────────────────────────────────────────┐    │
+│  │ HEURISTIC CAPTURED:                                       │    │
+│  │ "às vezes" = variação não documentada                     │    │
+│  │                                                           │    │
+│  │ Automatable? ⚠️ Needs clarification                       │    │
+│  │ → Added to GAP REPORT                                     │    │
+│  └──────────────────────────────────────────────────────────┘    │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**Decision Rule Format:**
+
+```
+IF [condition extracted from transcript]
+THEN [action A]
+ELSE IF [another condition]
+THEN [action B]
+ELSE [default action]
+
+Automatable: Yes | Partial | No
+Reason: [why]
+```
+
+**Heuristic Translation:**
+
+| Original Quote | Translation |
+|----------------|-------------|
+| "usually we wait 24h" | IF time_since_contact < 24h THEN wait ELSE proceed |
+| "depends on the client size" | IF client.revenue > X THEN priority = HIGH |
+| "when it feels right" | [NEEDS CLARIFICATION] → Add to gaps |
+
+---
+
+### Phase 6: Automation Analysis (PV_PM_001)
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 5: AUTOMATION ANALYSIS (PV_PM_001)                          │
+│  ─────────────────────────────────────────────────────────────── │
+│                                                                   │
+│  Para cada step:                                                  │
+│  ┌────────────┬───────────┬──────────────┬───────────┬─────────┐ │
+│  │ Step       │ Frequency │ Impact       │ Auto?     │ Decision│ │
+│  ├────────────┼───────────┼──────────────┼───────────┼─────────┤ │
+│  │ Abrir CRM  │ >4x/mês   │ Low          │ High      │AUTOMATE │ │
+│  │ Avaliar    │ >4x/mês   │ High         │ Medium    │HYBRID   │ │
+│  │ Ligar      │ 2x/mês    │ High         │ Low       │HUMAN    │ │
+│  │ Registrar  │ >4x/mês   │ Low          │ High      │AUTOMATE │ │
+│  └────────────┴───────────┴──────────────┴───────────┴─────────┘ │
+│                                                                   │
+│  ⚠️ VETO CHECK: Tem guardrails?                                   │
+│     □ Loop Prevention    □ Idempotency    □ Audit Trail          │
+│     □ Escape Route       □ Retry Logic    □ Rollback             │
+│                                                                   │
+│  Se não tem guardrails → VETO (não automatizar)                   │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**PV_PM_001 Decision Matrix:**
+
+| Frequency | Impact | Automatability | Decision |
+|-----------|--------|----------------|----------|
+| High (>4x/mo) | High | High | **AUTOMATE** |
+| High (>4x/mo) | High | Low | **DELEGATE** (specialized human) |
+| High (>4x/mo) | Low | High | **AUTOMATE** (efficiency) |
+| Low (<2x/mo) | High | Any | **KEEP_MANUAL** (risk doesn't justify) |
+| Low (<2x/mo) | Low | Any | **ELIMINATE** (question necessity) |
+| Any | Any | No guardrails | **VETO** (never automate without safeguards) |
+
+**Mandatory Guardrails:**
+
+| Guardrail | Apply To | Implementation |
+|-----------|----------|----------------|
+| **Loop Prevention** | Repetitive tasks | Deduplication + max_iterations |
+| **Idempotency** | Tasks with side-effects | Input hash → skip if identical |
+| **Audit Trail** | All tasks | Structured log per execution |
+| **Escape Route** | Critical tasks | manual_override flag available |
+| **Retry Logic** | External integrations | Exponential backoff (3 attempts) |
+| **Rollback** | Destructive tasks | Snapshot before execution |
+
+---
+
+### Phase 7: Quality Scoring (META-AXIOMAS)
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 6: META-AXIOMAS SCORING (10 Dimensions)                     │
+│  ─────────────────────────────────────────────────────────────── │
+│                                                                   │
+│  1. Truthfulness ........... 8/10  ✓                              │
+│  2. Coherence .............. 7/10  ✓                              │
+│  3. Strategic Alignment .... 8/10  ✓                              │
+│  4. Operational Excellence . 6/10  ✓                              │
+│  5. Innovation Capacity .... 5/10  ⚠ (below 6, flag)              │
+│  6. Risk Management ........ 7/10  ✓                              │
+│  7. Resource Optimization .. 7/10  ✓                              │
+│  8. Stakeholder Value ...... 8/10  ✓                              │
+│  9. Sustainability ......... 7/10  ✓                              │
+│  10. Adaptability .......... 6/10  ✓                              │
+│                                                                   │
+│  OVERALL: 6.9/10  ⚠️ Below 7.0 threshold → REVIEW needed          │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**META-AXIOMAS Dimensions:**
+
+| # | Dimension | Weight | Threshold | Description |
+|---|-----------|--------|-----------|-------------|
+| 1 | **Truthfulness** | 1.0 | 7.0 | Process described accurately? (VETO if <7) |
+| 2 | **Coherence** | 0.9 | 6.0 | Steps align logically? |
+| 3 | **Strategic Alignment** | 0.9 | 6.0 | Serves business goals? |
+| 4 | **Operational Excellence** | 0.8 | 6.0 | Is it efficient? |
+| 5 | **Innovation Capacity** | 0.7 | 5.0 | Can it evolve? |
+| 6 | **Risk Management** | 0.8 | 6.0 | Risks addressed? |
+| 7 | **Resource Optimization** | 0.8 | 6.0 | Is it lean? |
+| 8 | **Stakeholder Value** | 0.7 | 6.0 | Serves users? |
+| 9 | **Sustainability** | 0.7 | 6.0 | Maintainable? |
+| 10 | **Adaptability** | 0.6 | 5.0 | Handles change? |
+
+**Thresholds:**
+- Overall Score: ≥7.0 to proceed
+- Per Dimension: ≥6.0 (except Innovation/Adaptability: ≥5.0)
+- Truthfulness: VETO if <7.0
+
+---
+
+### Phase 8: Squad Blueprint Generation
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 7: SQUAD BLUEPRINT GENERATION                               │
+│  ─────────────────────────────────────────────────────────────── │
+│                                                                   │
+│  AGENTS:                                                          │
+│  ┌─────────────────────────────────────────────────────────────┐ │
+│  │ @customer-process-orchestrator  │ Orchestrates full flow    │ │
+│  │ @crm-worker                     │ CRM queries (Worker)      │ │
+│  │ @customer-analyst               │ Analysis (Agent)          │ │
+│  │ @customer-service               │ Human decisions (Human)   │ │
+│  └─────────────────────────────────────────────────────────────┘ │
+│                                                                   │
+│  TASKS:                                                           │
+│  ┌─────────────────────────────────────────────────────────────┐ │
+│  │ fetch-customer-data    │ @crm-worker      │ Worker          │ │
+│  │ analyze-customer       │ @customer-analyst │ Agent           │ │
+│  │ decide-priority        │ @customer-service │ Hybrid (elicit) │ │
+│  │ contact-customer       │ @customer-service │ Human           │ │
+│  └─────────────────────────────────────────────────────────────┘ │
+│                                                                   │
+│  CHECKPOINTS:                                                     │
+│  ┌─────────────────────────────────────────────────────────────┐ │
+│  │ CP1: After analyze-customer  │ Human validates analysis     │ │
+│  │ CP2: After decide-priority   │ Human confirms decision      │ │
+│  └─────────────────────────────────────────────────────────────┘ │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**Blueprint Structure:**
+
+```yaml
+squad_blueprint:
+  name: "[process-name]"
+  version: "1.0.0"
+
+  agents:
+    - name: "@[process]-orchestrator"
+      type: Orchestrator
+      responsibility: "Orchestrate complete flow"
+      steps: all
+
+    - name: "@[function]-worker"
+      type: Worker
+      responsibility: "[specific function]"
+      steps: [1, 4, 7]
+
+    - name: "@[function]-analyst"
+      type: Agent
+      responsibility: "[analysis function]"
+      steps: [2, 5]
+
+    - name: "@[function]-service"
+      type: Human
+      responsibility: "[human decisions]"
+      steps: [3, 6]
+
+  tasks:
+    - name: "[task-name]"
+      agent: "@[agent]"
+      execution_type: Worker | Agent | Hybrid | Human
+      elicit: true | false
+      checkpoint: true | false
+
+  checkpoints:
+    - id: CP1
+      after_task: "[task-name]"
+      type: validation | approval | veto
+      validator: human | @agent
+      criteria: "[approval criteria]"
+      action_on_reject: "[what happens]"
+
+  workflow:
+    name: "[process-name]-workflow"
+    trigger: "[what starts it]"
+    phases:
+      - name: "[phase-name]"
+        agent: "@[agent]"
+        tasks: ["task1", "task2"]
+        checkpoint: CP1 | null
+    handoff:
+      pattern: HO-TP-002
+      data_transferred: [artifacts]
+```
+
+---
+
+### Phase 9: Gap Analysis
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 8: GAP ANALYSIS                                             │
+│  ─────────────────────────────────────────────────────────────── │
+│                                                                   │
+│  ┌─────────────────────────────────────────────────────────────┐ │
+│  │ GAPS IDENTIFIED:                                            │ │
+│  │                                                             │ │
+│  │ 1. [INFERRED] Critério de "cliente premium" não definido    │ │
+│  │    → Pergunta: Qual o threshold de faturamento?             │ │
+│  │                                                             │ │
+│  │ 2. [MISSING] Tempo máximo de resposta não mencionado        │ │
+│  │    → Pergunta: Qual SLA para cada tier?                     │ │
+│  │                                                             │ │
+│  │ 3. [RED FLAG] "João sabe fazer isso" mencionado             │ │
+│  │    → Single point of failure identificado                   │ │
+│  │    → Pergunta: Quem mais pode executar?                     │ │
+│  └─────────────────────────────────────────────────────────────┘ │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**Gap Categories:**
+
+| Category | Marker | Impact | Action |
+|----------|--------|--------|--------|
+| **Inferred** | `[INFERRED]` | Medium | Validate with process owner |
+| **Missing** | `[MISSING]` | High | Request clarification |
+| **Ambiguous** | `[AMBIGUOUS]` | Medium | List interpretations, ask |
+| **Red Flag** | `[RED FLAG]` | High | Document risk, recommend mitigation |
+
+**Gap Report Format:**
+
+```markdown
+## Gap Report
+
+### Critical Gaps (Block Automation)
+| Gap | Impact | Question | Priority |
+|-----|--------|----------|----------|
+| [description] | Blocks automation | [question] | High |
+
+### Medium Gaps (Reduce Quality)
+| Gap | Impact | Question | Priority |
+|-----|--------|----------|----------|
+| [description] | Reduces quality | [question] | Medium |
+
+### Red Flags (Risks)
+| Flag | Risk | Mitigation | Owner |
+|------|------|------------|-------|
+| [description] | [risk] | [recommendation] | [who] |
+```
+
+---
+
+### Phase 10: Document Assembly & Validation
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  STEP 9: DOCUMENT ASSEMBLY (SC-PE-001)                            │
+│  ─────────────────────────────────────────────────────────────── │
+│                                                                   │
+│  OUTPUT FILES:                                                    │
+│                                                                   │
+│  📄 sop_document.md (11 parts)                                    │
+│     ├── Part 1: Process Identification                            │
+│     ├── Part 2: Stakeholders & Executors                          │
+│     ├── Part 3: Data & Systems                                    │
+│     ├── Part 4: Process Flow (with Mermaid diagram)               │
+│     ├── Part 5: Quality & Validation                              │
+│     ├── Part 6: Exceptions & Troubleshooting                      │
+│     ├── Part 7: Automation Analysis                               │
+│     ├── Part 8: AIOX Squad Blueprint                              │
+│     ├── Part 9: Gaps & Next Steps                                 │
+│     ├── Part 10: Glossary                                         │
+│     └── Part 11: Extraction Metadata                              │
+│                                                                   │
+│  📄 squad_blueprint.yaml                                          │
+│  📄 gap_report.md                                                 │
+│  📄 automation_analysis.md                                        │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**Validation Checklist (SC-CK-001):**
+
+| Category | Criteria | Threshold |
+|----------|----------|-----------|
+| Structure | All 11 parts present | 11/11 |
+| Task Anatomy | 8 fields per step | 100% |
+| Executor Classification | All steps assigned | 100% |
+| Cognitive Analysis | All steps classified | 100% |
+| Decision Rules | All "if/depends" captured | 100% |
+| Guardrails | Defined for automated steps | 100% |
+| META-AXIOMAS | Overall score | ≥7.0 |
+| Squad Blueprint | Complete and valid | 100% |
+| Gaps | All documented with questions | 100% |
+
+**Validation Decision:**
+
+| Decision | Criteria | Next Step |
+|----------|----------|-----------|
+| **APPROVE** | All thresholds met | Handoff to create-squad |
+| **REVIEW** | Some thresholds below | Human review required |
+| **REJECT** | Critical gaps | Re-extraction needed |
+
+---
+
+### Phase 11: Handoff & Squad Creation
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│  HANDOFF → /AIOX:agents:squad-creator → *create-squad            │
+│  ─────────────────────────────────────────────────────────────── │
+│                                                                   │
+│  handoff:                                                         │
+│    to: "create-squad"                                             │
+│    data_transferred:                                              │
+│      - sop_document.md                                            │
+│      - squad_blueprint.yaml                                       │
+│      - gap_report.md                                              │
+│    validation: "Process owner confirms accuracy"                  │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+**Final Output Structure:**
+
+```
+squads/[process-name]/
+├── agents/
+│   ├── [process]-orchestrator.md
+│   ├── [function]-worker.md
+│   ├── [function]-analyst.md
+│   └── [function]-service.md
+├── tasks/
+│   ├── [task-1].md
+│   ├── [task-2].md
+│   └── [task-n].md
+├── workflows/
+│   └── [process]-workflow.yaml
+├── checklists/
+│   └── [process]-validation.md
+├── config/
+│   └── squad-config.yaml
+├── config.yaml
+└── README.md
+```
+
+---
+
+## Summary Table
+
+| Phase | Step | Input | Output | Actor |
+|-------|------|-------|--------|-------|
+| 1 | Data Source | Config | Normalized transcript | System |
+| 2 | Structure | Transcript | Steps, roles, tools | @squad-chief |
+| 3 | Task Anatomy | Steps | 8-field tasks | @squad-chief |
+| 4 | Cognitive | Steps | Executor types | @squad-chief |
+| 5 | Decision Rules | Transcript | IF/THEN rules | @squad-chief |
+| 6 | Automation | Steps | PV_PM_001 decisions | @squad-chief |
+| 7 | Quality | Process | META-AXIOMAS score | @squad-chief |
+| 8 | Blueprint | All above | Squad YAML | @squad-chief |
+| 9 | Gaps | All above | Gap report | @squad-chief |
+| 10 | Assembly | All above | SOP document | @squad-chief |
+| 11 | Validation | SOP | APPROVE/REVIEW/REJECT | Human + SC-CK-001 |
+| 12 | Handoff | Approved SOP | Working squad | @squad-chief |
+
+---
+
+## Quick Start
+
+```bash
+# 1. Configure data source
+# Edit: squads/squad-creator/config/squad-config.yaml
+# Set: data_sources.transcripts.active_source
+
+# 2. Run extraction
+/extract-sop
+
+# 3. Follow guided workflow
+*extract-sop
+
+# 4. Review outputs
+# - SOP document (11 parts)
+# - Squad blueprint
+# - Gap report
+
+# 5. Validate (human review)
+*validate-sop
+
+# 6. Create squad
+/AIOX:agents:squad-creator
+*create-squad
+```
+
+---
+
+## Related Documents
+
+| Document | Location | Purpose |
+|----------|----------|---------|
+| SOP Template | `templates/pop-extractor-prompt.md` | SC-PE-001 structure |
+| Extraction Task | `tasks/extract-sop.md` | Workflow definition |
+| Validation Checklist | `checklists/sop-validation.md` | Quality gates |
+| Agent Definition | `agents/squad-chief.md` | @squad-chief persona (includes SOP extraction) |
+| Squad Config | `config/squad-config.yaml` | Data sources & settings |
+
+---
+
+**Pattern Compliance:** SC-DOC-001 ✓ | SC-PE-001 ✓ | HO-TP-001 ✓ | PV_PM_001 ✓
+
+*Generated by Squad Creator Documentation System*